In the present-day analytical approach, information and insights are
seen everywhere. If you can gather/analyze them, then you can
benefit from them. The traditional methods of data collection and
analysis do not include all the insights in the text materials. Yet, such
materials hold vital information that can be extracted and analyzed,
and the insights can be used in decision-making. Examples of such
materials are social media streams – Tweet feeds, Facebook posts,
etc., presidential speeches, memos, and so on.
Sentimental analysis is the study of opinion (judgment), attitude
(viewpoint), and emotions (state of mind). It is a text classification tool
to determine whether the subject or objective of the source material.
As a data scientist or analyst, my interest in the topic of sentimental
analysis will be to utilize the classification ability to detect the polarity
of the society and use it to make an informed decision, in both
businesses and governments. Also, the interest of a data scientist will
include the advancement of artificial intelligence ability. If the
government understands the direction of the society, it can enact
policies that will positively impact the community. Businesses can, on
the other hand, plan effective marketing and financial predictions.
In sentimental analysis, grammar correctness is not the concern. Data
mining plays a vital role in removing noise from the data. The process
contributes more than 80% success of the whole analysis. In
sentimental analysis, data mining is used to reduce the size of the
words in the document, as not all the words are positively contributing
to the analysis. The stop words such as “is”, “a”, “the”, “for”, and so on,
for example, are not needed, therefore need to be removed before
creating the bag-of-words. Also, words with the same linguistic root
will be put together as one single word. “Representation” and
“represented” have the same linguistic root. Similarly, machine
learning plays a very important role in sentimental analysis to train the
system on the pattern of the text document. Machine learning
techniques are used to classify the features of the text materials.
The sentimental analysis technologies are the key to this type of
analysis. It is humanly impossible to analyze any text materials without
the use of technology such as IBM Watson Studio for R. Hence, the
role of technology in the sentimental analysis is inevitable.
It is so interesting and useful to see some of the impossibilities of the
past to become possible. Companies are leveraging these technologies
to improve businesses. But text materials are meant to be read and
understood. If the sentimental analysis does not put much emphasis
on grammatical correctness, much of the content will be lost.
